A Feature Selection Method for Tissue-Specific Alternative Polyadenylation Sites Data in Rice. Issue 5 (March 2019)

• Record Type: Journal Article
• Authors: Zou, Wenbing; Chen, Moliang; Li, Shuchao; Ye, Pengchao; Ji, Guoli
• Is Part Of: IOP conference series. Volume 242:Issue 5(2019)
• Publication Date: 2019-03
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/242/5/052035
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: The identification of tissue-specific alternative polyadenylation (tsAPA) sites contributes to the research on gene expression regulation and transcriptome diversity in rice. However, identifying the tsAPA sites in plants is difficult, because of the dispersion, variability, complexity of their features and the lack of related research. A hybrid feature selection algorithm called SRBT, based on the SVM-RFE and Boruta, was presented to identify the tsAPA sites in rice. In the experiment, the tsAPA sites data were adopted to reduce dimension with SRBT algorithm and then classified by the support vector machine (SVM). The results show that the proposed method can effectively extract important features and obtain a higher average prediction accuracy of 81%, compared with SVM-RFE, Boruta, GAFS, T-test and ReliefF. The SRBT works well in identifying the tsAPA sites, which offers an effective method for further analysis of the tsAPA in gene expression and transcription during the growth of rice.